DESCRIPTION: The translational repression of the phage MS2 replicase gene by its coat protein is one of the most widely studied and fruitful systems on the specificity of RNA binding by protein. The structures of the protein and RNA target are known, individually and in the complex. Both components are small, can be prepared in large quantity and their binding can be easily quantitated. A genetic system facilitates selection of protein variants with altered specificity. Finally, it is a member of a homologous family of proteins that bind a similar but significantly different set of ligands, allowing the structure of one protein to serve as the model for others. The crystal structure of the coat protein-RNA complex shows the two bases of the MS2 RNA target, a bulged A at -10 and a loop A at -4, are bound in a quasi symmetric way by four or five amino acid residues on each of the monomers of the active coat protein dimer. Experiments proposed here will systematically explore the contributions of each of the amino acid side chains to this these interactions, using site directed mutations to substitute carefully chosen alternate amino acids at these sites. The PI has shown that the coat protein dimer can be replaced by two monomers fused into a single polypeptide chain. This allows manipulation of each of the A-binding sites independently of the other as fused heterodimers. It has the further useful property of stabilizing mutant proteins that would not be stable as monomers. In previous work the PI has identified RNA binding sites and coat protein specificity determinants for the related phages MS2, GA, Q-beta and SP, and has selected variants that interconvert one protein to the specificity for another RNA. He now proposes to extend these studies to phage PP7, which is the most distantly related of this group and where there is no data to indicate that coat protein repression of replicase occurs. He will use the same two plasmid system as previously, where the coat protein gene (targeted for mutagenesis) is on one plasmid and a lacZ gene regulated by a phage translational operator on the other. Transformants are screened for loss of repression and only mutant proteins that are still able to assemble into capsids are chosen for further study. This screens out mutations that result in misfolding, degradation or other changes not relevant to RNA binding. Finally, in a major departure from previous work, the PI will initiate a program of in vitro evolution of the binding specificity of the MS2 coat protein. Selex will be used to find the new RNA structures that are optimal for binding to specific coat protein variants. The fused heterodimer construct will allow the two halves of the binding site to be varied independently. In vitro selection will be extended to the protein by the simultaneous use of site directed codon randomization, DNA shuffling PCR, and phage display. The RNA target will be held constant initially and immobilized through a sandwich with a partially complementary biotinylated oligo. These targets are chosen to be the preferred binding sites for other, well characterized RNA binding proteins that have some variant of the beta sheet RNA binding domain found in MS2 coat: U1A, HIV Tar and T4 RegA. These RNAs are significantly different in structure. Eventually, coevolution of both RNA and protein components will be used to explore the potential for RNA specificity inherent in this protein fold.